African Journal of Emergency Medicine (2015) 5, 127135

African Federation for Emergency Medicine

African Journal of Emergency Medicine

www.afjem.com
www.sciencedirect.com

CLINICAL REVIEW

Sepsis, severe sepsis, and septic shock: A review of the literature

Septicemie, septicemie grave et choc septique: etude bibliographique

Keegan Tupchong a, Alex Koyfman b,*
, Mark Foran a

a
Department of Emergency Medicine, New York University School of Medicine, Bellevue Hospital Center, New York, NY, USA
b
Division of Emergency Medicine, UT Southwestern Medical Center/Parkland Memorial Hospital, Dallas, TX, USA

Received 25 November 2013; revised 14 April 2014; accepted 25 May 2014; available online 1 August 2014

Sepsis represents a continuum of illness due to systemic inammation caused by an infection that requires prompt recognition and treatment. While sepsis is a signif-
icant cause of death worldwide, its mortality is believed to be disproportionately high in low- and middle-income countries (LMICs). Since 1992, its denition has
become standardized, and beginning in 2002, an international collaboration has produced a set of consensus guidelines on the optimal management of septic patients.
Based on new evidence, signicant updates have been made since then. It is known that application of a bundled approach to patient care with the use of specic resus-
citation endpoints to guide therapy leads to signicant reductions in mortality from sepsis. However, it is also recognized that the implementation of such interventions
in LMICs is extremely challenging. Consequently, a body of literature on practical guidelines for sepsis in developing countries has emerged. This article provides a
review of the evidence for the best practice of sepsis management, with recommendations for resource-limited settings.

La septicemie represente un eventail de maladies dues a une inammation systemique provoquee par une infection qui necessite une identication et un traitement
rapide. Si la septicemie est une cause de deces importante dans le monde, la mortalite associee est consideree comme disproportionnellement elevee dans les pays a
faible et moyen revenus. Depuis 1992, sa denition sest standardisee, et a partir de 2002, une collaboration internationale a produit un ensemble de lignes directrices
consensuelles sur la prise en charge optimale des patients souffrant de septicemie. Sur la base de nouvelles donnees factuelles, des mises a jour denvergure y ont depuis
ete apportees. On sait que lapplication dune approche globale a la prise en charge des patients, combinee au recours a des criteres de reanimation visant a guider la
therapie, permet dobtenir une reduction signicative de la mortalite liee a la septicemie. Il est cependant egalement reconnu que la mise en uvre de telles interventions
dans les pays a faible et moyen revenu est extremement difcile. Par consequent, un certain nombre de publications sur les directives pratiques relatives a la prise en
charge de la septicemie dans les pays en voie de developpement sont apparues. Cet article propose un examen des elements probants sur une meilleure pratique de la
gestion de la septicemie, ainsi que des recommandations pour les milieux a ressources limitees.

African relevance
 Sepsis has an especially high mortality in Africa.
 Rapid recognition of sepsis and a protocoled approach to
management save lives.
 Tailored recommendations for resource-limited settings
offer a practical approach to sepsis care.
Introduction
In 1992, sepsis was formally dened as the presence of both
suspected infection and two of the four criteria of the systemic
inammatory response syndrome (SIRS) (Tables 1 and 2).1,2
* Correspondence to Alex Koyfman. akoyfman8@gmail.com
Peer review under responsibility of African Federation for Emergency Medicine.
Production and hosting by Elsevier
http://dx.doi.org/10.1016/j.afjem.2014.05.004
2211-419X 2014 Production and hosting by Elsevier B.V. on behalf of African Federation for Emergency Medicine.
Since then, additional terminology has emerged. Sepsis compli-
cated by organ dysfunction is referred to as severe sepsis,
while sepsis complicated by hypotension refractory to ade-
quate volume resuscitation in the absence of an alternate cause
has been termed septic shock.1 The clinical signicance of
the sepsis spectrum of illness cannot be understated. In the
US alone, the incidence of severe sepsis is over 700,000 annu-
ally with an estimated 30% mortality.3 This is estimated to
represent over 450,000 emergency centre (EC) visits per year.
4
While some research has been devoted to the study of sepsis
in developing countries, its epidemiology in these countries
remains poorly described.5,6 Despite this, the morbidity and
mortality of sepsis in low- and middle-income countries
(LMICs) are believed to be disproportionately high, given
environmental degradation, widespread malnutrition, and
higher rates of bacterial, parasitic, and HIV infection.6,7 In
an effort to reduce the risk of death from sepsis, the Surviving
Sepsis Campaign (SSC) was initiated in 2002 from the collab-
oration of the European Society of Intensive Care Medicine
(ESICM), the International Sepsis Forum (ISF), and the
Society of Critical Care Medicine (SSCM). In 2004, the SSC
128 K. Tupchong et al.

SS, this article will primarily discuss SS/SS. A particular focus

Table 1 SIRS criteria.1
will be on providing care in resource-limited settings with
Presence of two or more of the following. tailored recommendations. Of note, this review pertains
1. Temperature  >38 C (100.4 F) or specically to adults and not to children. See Tables 1 and 2
 <36 C (96.8 F) for denitions.
2. Heart rate  >90/min
3. Respiratory rate  >20/min or Epidemiology
 PaCO2 < 32 mmHg
4. White blood cell count  >12,000/lL or
 <4000/lL Despite the documented impact of sepsis in developed coun-
tries, literature on its incidence, prevalence, and mortality in
developing countries is sparse.10 What is recognized, however,
is that the global burden of sepsis lies in LMICs. As a surro-
Table 2 Diagnostic criteria for sepsis.1 WBC, white blood gate marker for sepsis, over 90% of worldwide deaths due to
cell; SBP, systolic blood pressure; MAP, mean arterial pressure. pneumonia, meningitis, and other infections occur in less
Table adapted from Levy et al. (2003).1 developed nations.6,11 Globally, an estimated 70% of the 9
Infection (documented or suspected) and some of the following. million annual neonatal and infant deaths are attributable to
sepsis, and more than half of these occur in Asia and Sub-
Classication Variables
Saharan Africa.11,12
General Fever (>38.3 C)
Hypothermia (core temperature < 36 C)
Pathophysiology
Heart rate > 90/min or more than two SD
above the normal value for age
Tachypnea During infection, offending microbes interact with the host
Altered mental status immune system producing a downstream inammatory cas-
Signicant oedema or positive uid balance cade involving cytokines and other mediators, which in turn
Hyperglycaemia triggers a systemic response. The resultant effects include
Inammatory Leukocytosis vasodilation, increased vascular permeability, myocardial
Leukopenia depression, and impairment of the coagulation cascade,
Normal WBC count with greater than 10% resulting in global imbalance of systemic oxygen supply
bands and demand, and a procoagulant state. During the late stage
Plasma C-reactive protein > 2 SD above the of sepsis, immunosuppression predominates, leading to
normal value
multiorgan dysfunction and further clinical deterioration.13
Plasma procalcitonin > 2 SD above the
normal value
Clinical assessment
Hemodynamic Hypotension (SBP < 90 mmHg,
MAP < 70 mmHg, or an SBP
decrease > 40 mmHg in adults or < 2 SD During history taking, the focus should be on detecting risk
below normal for age) factors for infection (such as immunosuppression), the pres-
ence of infection, and if suspected, the most likely sources.
Organ dysfunction Creatinine increase
Caution should be advised in geriatric patients, as they may
Coagulopathy
Hypoxaemia not be able to communicate traditional symptoms (e.g., dys-
Ileus uria in occult urinary tract infections). The physical examina-
Oliguria tion should be used to identify possible foci of source control.
Thrombocytopenia A critical action at this point is the measurement, documenta-
Hyperbilirubinemia tion, and evaluation of vital signs, including temperature,
Tissue perfusion Hyperlactatemia
blood pressure (BP), heart rate (HR), respiratory rate (RR)
Decreased capillary rell or mottling and oxygen saturation (if below 90% then supplemental
oxygen should be immediately applied). Repeated recording
of these parameters will be used to gauge clinical improvement
produced the Surviving Sepsis Campaign guidelines for or deterioration and trigger specic interventions (see below).
management of severe sepsis and septic shock, one of the Consistently analysing the vital signs for the presence of SIRS
most recognized consensus statements regarding the treatment criteria in any possible patient with sepsis will aid in the early
of sepsis (most recently updated in 2012).8 In many countries, recognition of critical illness.
it is held to be the standard of care.7 As 50% of hospital Importantly, vital sign derangements may be absent early
admissions occur through the ED, there is a signicant oppor- on and in elderly patients.14,15 Specic physical exam ndings
tunity to improve outcomes.9 This review will discuss the that are predictive of sources of infection include indwelling
epidemiology, pathophysiology, and diagnostic and therapeu- devices (e.g., intravascular or urinary catheters), rales,
tic approach to patients with sepsis, severe sepsis, and septic abdominal tenderness, and evidence of CNS infection.16,17 A
shock in the ED and other acute care settings. It is important cardiovascular and volume status assessment, including
to note that as international guidelines focus on the evaluation auscultation, mucous membranes, skin colour and turgor,
and management of patients with severe sepsis and septic peripheral pulses, capillary rell and oedema should be
shock (SS/SS) as opposed to sepsis without evidence of SS/ undertaken at this stage as well.
Sepsis, severe sepsis, and septic shock
In the absence of automated devices to track vital signs, it is
imperative to keep a written log and to have early recognition
of SIRS. As more advanced and invasive equipment used to
gauge response to therapy (see below) may be absent in
LMICs, changes in BP, HR, RR and oxygen saturation to
interventions should be used as indicators of clinical status.
Diagnostic studies
Initial laboratory and radiographic testing is aimed at locating
a source of infection and identifying evidence of organ dys-
function. At the discretion of the examiner, common labora-
tory studies include a complete blood count (white blood cell
count including a differential of subtypes and measure of
bands, haemoglobin and haematocrit, platelets), chemistries
(electrolytes, bicarbonate, creatinine, glucose), prothrombin
time (PT)/international normalized ratio (INR), liver transam-
inases, bilirubin, and either arterial or venous blood gas anal-
ysis with the inclusion of a serum lactate level. Urinalyses are
of high-yield, particularly in patients older than 65.18 Chest
radiography is often sought to identify sources of pulmonary
infection and causes of respiratory distress. Importantly, all
patients with possible sepsis should have blood cultures drawn
as they may narrow antibiotic choices and uncover reasons for
treatment failure.8 As the volume of blood is more important
for culture yield than the number of blood cultures, ideally,
20 mL of blood should be collected.19 In general, cultures
should be obtained from every possible source of infection
(e.g., blood, urine, sputum, wounds, catheters, cerebrospinal
uid) deemed likely in the clinical scenario.
If only select laboratory tests are available, preference
should be given to those that may indicate a source of infec-
tion, consequently identifying a target for therapy. Blood cul-
tures should be obtained whenever possible.
Treatment and management
General approach
Prompt recognition of the septic patient is critical, and early
localization along the sepsis spectrum of illness (sepsis, severe
sepsis, septic shock) helps to dene the early goals of manage-
ment. A key distinction should be made between sepsis and
severe sepsis/septic shock (SS/SS), the latter of which are the
focus of the Surviving Sepsis Campaign guidelines. It is recom-
mended that patients with SS/SS undergo a protocol-driven
approach to treatment using quantitative resuscitation.8 This
is the concept of using explicit, predened physiological or lab-
oratory resuscitation endpoints to guide management.20 Its
international recognition was sparked by a landmark study
by Rivers in 2001, which utilized a goal-directed approach to
therapy of SS/SS patients (early goal-directed therapy, EGDT)
and demonstrated a 16% absolute reduction in in-hospital
mortality (number needed to treat = 6.25).21 A 2008 meta-
analysis provided further support for the use of early and
quantitative resuscitation as it showed a survival benet in het-
erogeneous populations with sepsis.22 Of note, however, the
ProCESS trial, a recent multicenter randomized trial of over
1300 patients demonstrated no mortality benet with the use
of protocol-based resuscitation (such as EGDT) over usual H2O. Values signicantly below this may be suggestive of
129
(provider-driven) care.23 It is critical to note, however, that
patients in the usual care group were aggressively resusci-
tated (in the rst 6 h of EC arrival: 4.4 L of intravenous uids,
58% had central venous catheters, 44% were placed on vaso-
pressors, 22% underwent mechanical ventilation), and in the
absence of this high level of usual care, protocol-based
resuscitation may still be advocated.
Some of the markers of resuscitation and their usage (e.g.,
central venous pressure, CVP; mean arterial pressure, MAP;
central venous oxygen saturation, ScvO2, serial lactate) will
be discussed below.
Patients with hypotension despite uid resuscitation and
those with evidence of end organ dysfunction (SS/SS) have
high mortalities and aggressive resuscitation or adherence to
protocoled, goal-oriented therapy may improve mortality.
Using as many of the available following techniques can alter
the course of illness. Upon recognition of SS/SS, a team of care
providers should be alerted and enlisted.
Initial resuscitation
The initial evaluation of patients with sepsis includes the rapid
establishment of multiple points of intravenous (IV) access
with anticipation for the need for uid resuscitation, antimi-
crobials, and possibly vasoactive medications (vasopressors).
As patients are often hypovolaemic, frequently suffering large
uid decits, immediate administration of 30 mL/kg (typically
2 L) of crystalloid is recommended.8,24 The goal is to restore
tissue perfusion, and volume resuscitation leads to increases
in cardiac output and systemic oxygen delivery.24 Patients with
hypotension (MAP < 65 mmHg) after a uid bolus, hyperlac-
tatemia (>4 mmol/L), signs of hypoperfusion or organ failure
are deemed to be in SS/SS. In such instances, transition to a
goal-directed approach to therapy is recommended by consen-
sus guidelines with measurable endpoints at the 3- and 6-h
marks.8 It is important to note that patients meeting criteria
for sepsis but not SS/SS may benet from a similar approach
to therapy. However, consensus guidelines advocating bundled
care pertain specically to SS/SS patients.8
In resource limited settings, administer at least 30 mL/kg
(typically 2 L) of crystalloid upon the recognition of sepsis,
as many patients are hypovolaemic.
Measures of intravascular volume
The rst protocoled step in the care of patients with SS/SS is
uid resuscitation with the goal of restoring intravascular vol-
ume (and presumptively, preload).8 There are several methods
to estimate volume status that are gaining support, such as
ultrasound-guided assessment of the respirophasic variation
of the inferior vena cava (IVC), and pulse-pressure variation
(PPV).25 However, the efcacy of these monitoring techniques
in improving clinical outcomes is believed to require further
study.8 As such, consensus guidelines continue to recommend
CVP measurement as a surrogate for preload despite a well-
known systemic review demonstrating a very poor relationship
between CVP and uid responsiveness.8,26 Consequently, the
rst resuscitation endpoint in SS/SS according to the Surviving
Sepsis Campaign guidelines is a target CVP of 812 cm
130
hypovolaemia and the potential need for additional uid resus-
citation. It is worth highlighting here that utilizing CVP requires
the placement of a central venous catheter above the diaphragm.
In the resource-limited setting, be aggressive in empirically
uid loading patients as they may require more than 46 L of
crystalloid in the rst 6 h of care.21 This intervention alone has
been shown to reduce mortality.27 As shown in the ProCESS
trial, aggressive uid resuscitation (an average of 4.4 L in the
rst 6 h in this study) should be the norm.23 In the absence
of advanced modalities to estimate uid responsiveness, check
jugular venous distension and listen for crackles as evidence of
uid overload.28
Ongoing uid administration
While highly variable by case, patients in the ProCESS trial
usual care arm received an average of 4.3 3.9 L of intrave-
nous uid from hours 6 to 72 (after initial resuscitation).23
In resource-limited settings, use markers of hypoperfusion
such as hypotension, lactate clearance, and urine output (see
below) to guide the rate of uid administration following the
initial resuscitation period. While variable, this may be in
excess of 8 L over the following 72 h.
Fluid selection
The results of numerous trials support the preferential use of
crystalloid, the avoidance of hydroxyethyl starches (HES),
and the select use of albumin only after large quantities of crys-
talloid administration. Crystalloid has repeatedly shown to be
cheaper, and has either no mortality difference or a lower mor-
tality in comparison HES, along with a decreased need for
renal replacement therapy (RRT) and is strongly recom-
mended.8,29,30 Additionally, it is important to note that while
there are emerging data on the potential benets of using bal-
anced crystalloid solutions (e.g., Ringers lactate) as opposed
to chloride-rich uids (e.g., normal saline), the primary consid-
eration in LMICs should be on ensuring adequate crystalloid
volume resuscitation regardless of the type.31,32 It is important
to reiterate here that this review focuses on adult patients, as a
study of over 3000 children in shock with febrile illnesses who
were randomized to uid boluses vs. no boluses demonstrated
an increased mortality in the uid bolus group.33 The applica-
bility of this trial to adult patients has yet to be established.
In resource-limited settings, use crystalloids liberally. The
type of crystalloid infusion is not as important as the volume
given.
Vasopressors
The second resuscitation endpoint in SS/SS is the establish-
ment of a MAP of at least 65 mmHg. In SS/SS, there is an
imbalance between vasodilation and vasoconstriction. Accord-
ingly, vasoactive medications are often required in septic
patients with hypotension (MAP < 65 mmHg). Their purpose
is to restore adequate arterial pressures and blood ow to vital
organs. Below MAPs of 60 mmHg, coronary, renal and CNS
autoregulation are lost and organ blood ow has a linear rela-
tionship with pressure.34 However, tissue perfusion can be
maintained at MAPs as low as 65 mmHg.35 As such, when
K. Tupchong et al.
using vasopressors, targeting a MAP P 65 mmHg is an
accepted clinical end point for most patients.8 An important
caveat is that higher MAPs may be required in patients with
chronic hypertension and lower MAPs may be tolerated in
young patients.8 Ideally, uid resuscitation (CVP P 8 cm
H2O) should occur prior to the use of vasopressors, however,
patients in shock may require them to be started sooner.8
There are several options for choice of vasopressor agent.
In the Rivers study, dopamine and norepinephrine were
used.21 In a recent, multicenter trial comparing norepinephrine
and dopamine, both were deemed equally efcacious at revers-
ing hypotension and there was no signicant difference in mor-
tality, though there were more arrhythmias in the dopamine
group.36 The authors concluded that norepinephrine is
superior in the setting of sepsis.36
In patients with refractory hypotension despite uid resus-
citation and norepinephrine administration, vasopressin may
be added at low doses of 0.03 units per minute and has been
shown to have benecial hemodynamic and renal function
effects.37 Epinephrine and phenylephrine are alternative addi-
tions to norepinephrine.8
The newest consensus guidelines recommend norepineph-
rine as the rst-line vasopressor, while epinephrine and vaso-
pressin are possible additions. Dopamine is recommended
against being used except in highly select circumstances, such
as relative bradycardia.8
In resource-limited settings, vasopressors should be initi-
ated once patients are believed to be adequately uid resusci-
tated. If available, norepinephrine is preferred, followed by
epinephrine and/or vasopressin. Dopamine and phenylephrine
may be used if no other agent is available.
Blood products and inotropes
The third resuscitation endpoint for the treatment of SS/SS is a
SCVO2 of at least 70%, as a measure of the balance between
tissue oxygen delivery and consumption. There are several
causes of low central venous oxygen saturations (SCV-
O2 < 70%) in septic patients who have achieved adequate
uid resuscitation and adequate blood pressure. This may be
due to metabolically active tissue, arterial hypoxaemia, low
oxygen carrying capacity, or decreased cardiac output. The
degree of oxygen delivery can be improved by the transfusion
of packed red blood cells.21
Red blood cells should be transfused to a target haemoglo-
bin concentration of 79 g/dL.8 This differs from the Rivers
study, which used a threshold haematocrit of 30%.21 However,
while the optimal haemoglobin concentration in sepsis is not
specically known, results from the Transfusion Requirements
in Critical Care trial did not demonstrate increased mortality
when using a haemoglobin count of 79 g/dL compared to
1012 g/dL.38 Routine administration of fresh frozen plasma
to correct a coagulopathy should be avoided unless there is
active bleeding or a planned invasive procedure. Platelets
should be transfused below 5000/mm3 even in the absence of
bleeding, and may be considered between 5000 and 30,000/
mm3 if there is a signicant risk of bleeding. If the platelet count
is 50,000/mm3 or above then transfusion should only take place
in the setting of planned surgery or invasive procedures.8
If ScvO2 is less than 70% despite an appropriate haemoglo-
bin level (above), inotropic agents may be administered to
Sepsis, severe sepsis, and septic shock
augment cardiac output. Dobutamine (2.5 mcg/kg/min to
maximum 20 mcg/kg/min) is the agent of choice.21 Exceeding
this dose has not been shown to be effective.
In resource-limited settings, blood product transfusions
should be used sparingly, as indicated above. Inotropic ther-
apy can be initiated to augment cardiac output if evidence of
shock persists despite liberal uid administration, vasopressor
initiation, and blood product transfusion (if indicated).
As SCVO2 measurement requires a central venous catheter
(CVC), other measures of shock such as lactate and urine
output (see below) can be used as surrogate indicators.
Lactate
Lactate is an important marker of severe sepsis and elevated
levels (P4 mmol/L) are associated with a high mortality rate.39
A recent multicenter randomized trial by Jones compared lac-
tate clearance (P10%) with SCVO2 (P70%) as a quantitative
resuscitation endpoint and found it to be noninferior.40 This
suggests that measuring peripheral lactate levels instead of
SCVO2 (which requires a CVC) may spark a noninvasive path-
way of quantitative resuscitation. However, this possibility has
yet to be included in consensus guidelines, which currently
state a goal of lactate normalization.8 Importantly, SCVO2
and lactate need not be used to the exclusion of one another,
and when available, can be used concurrently.
In the absence of CVCs, checking serial lactate levels for
clearance can be used to guide therapy (above)
Urine output
Regardless of whether SCVO2 or lactate levels are used to guide
therapy, resuscitation within the rst 6 h includes the mainte-
nance of urine output P 0.5 mL/kg/h. This is a supplemental
endpoint to other markers of tissue perfusion, and low levels
suggest the need for more aggressive resuscitation.8
In resource-limited settings, urine output should be
measured for all patients. An indwelling urinary catheter can
be used, or a urine collection container. The time and amount
of each voiding episode should be recorded, similar to the vital
signs.
Antimicrobials
In the setting of SS/SS, it has been demonstrated that delayed
antibiotic administration is clearly associated with increased
mortality. Even prior to the onset of persistent or recurrent
hypotension, there is an 8% increase in mortality for each hour
of delay in antimicrobial administration.41 In another cohort
of patients undergoing EGDT, those who received appropriate
antibiotics less than one hour after triage had a 14% lower
mortality than those who did not.42 While timing is essential,
antibiotic choice is similarly important. Targeting the offend-
ing pathogens has outcome implications as well. In a prospec-
tive study, Gram-negative bacteraemia patients who
empirically received appropriate (pathogen susceptible)
antibiotics had a signicantly lower mortality than those
who empirically received inappropriate (pathogen not suscep-
tible) antibiotics (18% vs. 34%).43 Not surprisingly, patients
who receive inadequate therapy have a 10% increase in
hospital mortality.41 As such, it is recommended that
131
broad-spectrum antibiotics be given as soon as possible in sep-
sis, but always within the rst hour after recognition of SS/SS.8
While the feasibility of this requires further study, and is not yet
the international standard of care, the importance of adminis-
tering correct antibiotics as soon as possible cannot be
overstated.39
The choice of empirical antibacterial therapy varies signi-
cantly based on patient characteristics. Important consider-
ations include (but are not limited to): the most likely
source(s) of infection, recent antibiotics (last 3 months), recent
healthcare exposure (e.g., hospitalization), underlying chronic
disease, local pathogens and drug resistance.8 Patients with
recent antibiotic exposure have an elevated incidence of
high-risk infections such as methicillin-resistant Staphylococ-
cus aureus and Pseudomonas aeruginosa, whereas those with
recent healthcare exposure have increased likelihood of being
colonized with ESBL-producing bacteria.44 Importantly, the
most common pathogens in SS/SS among hospitalized patients
in the descending order are Gram-positive bacteria, Gram-neg-
ative bacteria, and mixed bacterial microorganisms.8 In all
cases, antibacterial therapy should be guided by the most likely
sources of infection along with local antibiograms.
In addition to bacteria, fungi have emerged worldwide as
growing causes of infection, particularly in the hospital setting,
and Candida is the fourth most common agent identied in
blood cultures of septic patients. Despite this, it is estimated
that fungi are responsible for only 5% of all cases of SS/SS,
and therefore, the routine administration of antifungals is
not recommended. Rather, they should be reserved for select
patients with known Candida colonization at multiple sites,
impaired physiologic barriers (recurrent gastrointestinal perfo-
rations or anastomotic leakages, acute necrotizing pancreatitis,
chemotherapy-induced mucositis, vascular access devices)
immunocompromised states (neutropenic cancer patients or
transplant recipients), or failure to improve while receiving
antibiotics.45 First-line agents are azoles or echinocandins.46
In resource-limited settings, rapid administration of appro-
priate antibiotics is a cornerstone to sepsis care and has a time-
sensitive effect on mortality. Antibiotics should be started as
soon as possible (ideally within the rst hour of the recognition
of sepsis). Important patient characteristics include recent use
of antibiotics and healthcare exposure. Local or regional anti-
biotic guidelines are preferred and should be used to choose
the most appropriate agent(s). Consider the addition of an
antiviral agent in select cases.28
Malaria
In Sub-Saharan Africa, malaria co-infection with sepsis is
common in children. In a study of over 3000 children in
Uganda, Kenya and Tanzania with febrile illnesses, 57% were
found to have malarial infection.33 In another study of chil-
dren with malaria, 6% had concomitant invasive bacterial
infection along with associated increased mortality.47 In
adults, however, the co-infection rate is less clear, with a recent
study of septic adults in Uganda, a malaria-endemic area,
demonstrating a malarial prevalence of only 4%.48
As there are many forms of rapid diagnostic tests (RDTs)
for malarial antigens (e.g., histidine-rich protein 2, Plasmodium
lactate dehydrogenase) now in use in endemic areas, parasito-
logical conrmation has risen from 20% to 47% from 2005
132
to 2011.49 Despite this, RDTs do not have high enough nega-
tive predictive values to support withholding treatment in the
presence of severe illness, and presumptive antimalarial treat-
ment for febrile illnesses has long been recommended.5054
In resource-limited settings, while there is insufcient
evidence to recommend empiric antimalarial treatment in all
adult patients with SS/SS, if there is clinical concern for
concomitant malarial infection, empiric treatment should be
started and guided by local resistance patterns.
Source control
Removing the source of infection is important as well and is
termed source control. It includes interventions such as uid Endotracheal intubation and mechanical ventilation can help
drainage, debridement of soft tissues, device removal, and
other denitive measures to restore function. The physical
removal of an ongoing locus of infection should be identied
or excluded as soon as possible as it can rapidly alter the
course of sepsis.8,55 Interventions should start with the least
invasive method (e.g., percutaneous as opposed to surgical)
when possible. When vascular catheters are left indwelling,
blood cultures should be sent from them. Indwelling urinary
catheters should be replaced and a urine culture should be sent
from the new catheter.17
In resource-limited settings, remove the source of infection
whenever possible, particularly if it can be drained or
debrided.11
Corticosteroids
SS/SS may be associated with a relative adrenal insufciency.56
As such, several studies have attempted to determine whether
the routine administration of corticosteroids may aid the reso-
lution of shock and improve overall mortality. As anti-inam-
matory agents, it is theorized that corticosteroids may reduce
the tissue damage caused by cytokines and neutrophils.13 Early
on, corticosteroids were not shown to be benecial in SS/SS
and in fact, potentially worsened outcomes.57 However, a later
study by Annane et al demonstrated that patients with vaso-
pressor-unresponsive septic shock developed shock reversal
and mortality reduction when treated with low-dose hydrocor-
tisone and udrocortisone.58 A follow-up, large multicenter
trial (CORTICUS), which included septic shock patients
who responded to vasopressors, did not show a survival benet
for patients receiving low-dose steroids.59 As such, consensus
guidelines recommend against the routine use of hydrocorti-
sone if IV uids and vasopressors can restore hemodynamic
stability. In the event of vasopressor-unresponsive shock,
hydrocortisone at a dose of 200 mg IV daily is suggested.8
In resource-limited settings, steroids, when available,
should be reserved for persistent septic shock following ade-
quate uid resuscitation and administration of vasopressors.
Glucose
While it was previously believed that tight glycaemic control
(blood glucose 80110 mg/dL) was benecial, data from the
NICE-SUGAR trial have demonstrated increased mortality
with this approach compared to conventional glucose control
(blood glucose 6 180 mg/dL).60 It is now recommended that
K. Tupchong et al.
a protocoled approach to insulin dosing begins only after
two blood glucose measurements > 180 mg/dL.8
In resource-limited settings, check blood glucose levels on
all patients. In the presence of mild hyperglycaemia, without
close monitoring, be cautious about reducing glucose levels,
as precipitating hypoglycaemia may be more harmful.
Intubation and mechanical ventilation
In septic patients, the development of shock in the rst 24 h is
the best predictor of the need for mechanical ventilation, and
half of all patients with this severity of illnesses go on to
develop acute respiratory distress syndrome (ARDS).61
by decreasing work of breathing and improving oxygen
delivery.8,21 When required, the decision should be made early
on. There are several options for induction agents in rapid
sequence intubation (RSI) and there is debate as to which is
the most favoured.
Etomidate is a commonly used induction agent in the Uni-
ted States. Its use in sepsis has become a focus of discussion,
particularly in recent years as the CORTICUS study showed
that patients receiving etomidate had a 17% increased rate
of adrenal suppression and increased risk of death at 28 days
(4045% vs. 3032%).59 However, since that time, several pro-
spective and retrospective studies have not shown a statistically
signicant increase in mortality with the use of etomidate.6265
As such, there are no denitive data at this time preventing the
use of etomidate in SS/SS. There are, however, advantages to
other agents such as ketamine, which has drawn attention due
to its sympathomimetic effects on heart rate, blood pressure,
and cardiac output.66,67 Presently, the choice of induction
agent is clinician-dependent.
The inammatory response to sepsis can cause lung injury
and the development of ARDS. In a landmark multicenter
randomized trial of lung-injured patients, those who under-
went mechanical ventilation with lung-protective ventilation
(low tidal volumes of 6 mL/kg of predicted body weight and
low plateau pressures 6 30 cm H2O) had a 9% absolute reduc-
tion in mortality.68 As such, this is the preferred ventilation
strategy. Following intubation, fractions of inspired oxygen
(FiO2) should be reduced as soon as possible to prevent oxygen
toxicity. Maintaining low levels of positive end-expiratory
pressure (PEEP) can prevent extensive alveolar collapse at
end-expiration.8,41,61 Additionally, the head of the bed should
be elevated 3045 as it has prospectively been shown to reduce
the rate of ventilator-associated pneumonia.69
In resource-limited settings, use pulse oximetry routinely
and keep oxygen levels above 90% by administering supple-
mental oxygen.
Summary
Consensus guidelines exist with specic recommendations for a
bundled approach to the treatment of septic patients, and in
particular, those with SS/SS. The concept of quantitative
resuscitation has gained support in recent years and has helped
standardize the approach to the care of sepsis. What is
included above is the inclusion of ideal practices. Key points
include:
Sepsis, severe sepsis, and septic shock 133

Table 3 Sepsis interventions in resource-limited settings.11 Table adapted from Dunser et al. (2012).11
Resource-limited interventions
Category Intervention
Circulation  Target adequate tissue perfusion as the primary endpoint of resuscitation. Additionally, aim for a blood
pressure > 90 mmHg
 When there is evidence of hypoperfusion, infuse crystalloid uids aggressively and continue liberal infusions for 24
48 h. Patients may require more than 4 L during the rst 24 h
 Preferentially use norepinephrine in patients with hypoperfusion despite liberal uid resuscitation. Monitor blood
pressure and heart rate frequently
 Reserve steroids for patients requiring escalating doses of vasopressors
Oxygenation  Apply oxygen with goal oxygen saturation > 90%. Administer empiric oxygen in patients with SS/SS in the absence of
a pulse oximeter
Antimicrobial therapy  Administer antimicrobials within 1 h of recognizing sepsis
 Choose antimicrobials with a high likelihood to be active against the suspected pathogens
Diagnosis  Gather a detailed patient history and perform a thorough examination, seeking to identify the source of infection
 Use imaging techniques when available
 Obtain cultures from every likely source of infection and test for antimicrobial sensitivities
Source control  Drain or debride the source of infection when possible
 Remove any foreign body or device that may be the source of infection

 Rapid screening using SIRS criteria and vigilant recogni-
tion of the sepsis spectrum of illness.
 Continually repeated vital signs.
 Supplemental oxygen for oxygen saturation levels < 90%.
 Immediate intravenous access at multiple sites.
 Fluid resuscitation using P 30 mL/kg of crystalloid fol-
lowed by intravascular volume assessment.
 Maintenance of MAP P 65 mmHg and the use of vasopres-
sors (norepinephrine rst line) if necessary.
 The use of blood products (haemoglobin 79 g/dL) and/or
dobutamine to keep SCVO2 P 70% (when available).
 Measurement and normalization of lactate levels.
 Cultures obtained from every suspected source of infection.
 Administration of broad-spectrum antibiotics as quickly as
possible (ideally within the rst hour).
 Early source control.
 Avoidance of corticosteroids except in vasopressor-unre-
sponsive shock.
 Conventional glucose control 6 180 mg/dL.
 Ventilation with low tidal volumes (6 mL/kg).
 Continued reassessment of clinical status and revaluation of
treatment modalities.
have been generated in their absence. They highlight the high
rate of HIV and other potentially contributing infectious dis-
eases (such as malaria and typhoid fever), regional variation
in bacteriology, and emphasize the importance of rapid recog-
nition of sepsis and the need for additional training and
research.6,11,75 Table 3 is a summary of a systematic review of
best-practice interventions tailored to resource-limited settings.
Future directions
Much is needed in the way of research pertaining to the care of
septic patients in LMICs. Particular areas of focus include the
epidemiology of sepsis, availability of resources in non-tertiary
care settings, efcacy of bundled care and the relative efcacy
of specic targeted interventions, health care worker knowl-
edge and training programs, and cost-benet analysis of each
of these components.
Conicts of interest
The authors declare no conict of interest.

African relevance
While the Surviving Sepsis Campaign guidelines have gained
international recognition and led to improved outcomes for
SS/SS patients in developed countries, many of the recommen-
dations cannot be implemented within LMICs due to a lack of
resources.7,70,71 For example, in many circumstances, one of
the most fundamental of needs, stafng for appropriate patient
monitoring, is inadequate, and there is poor access to higher
levels of care such as intensive care units.7,72 Furthermore,
there is debate about the transferability of some studies per-
formed in high-income countries.73 Nonetheless, there exists
evidence that a bundled approach to sepsis care in the develop-
ing world, focused on early uid and antibiotic administration
improves mortality in Africa.74
There are no accepted guidelines for SS/SS management in
LMICs. However, several evidence-based recommendations
References
1. Levy MM et al. 2001 SCCM/ESICM/ACCP/ATS/SIS interna-
tional sepsis denitions conference. Crit Care Med
2003;31(4):12506.
2. American College of Chest Physicians/Society of Critical Care
Medicine Consensus Conference: denitions for sepsis and organ
failure and guidelines for the use of innovative therapies in sepsis, .
Crit Care Med 1992;20(6):86474.
3. Angus DC et al. Epidemiology of severe sepsis in the United
States: analysis of incidence, outcome, and associated costs of
care. Crit Care Med 2001;29(7):130310.
4. Nguyen HB, Smith D. Sepsis in the 21st century: recent denitions
and therapeutic advances. Am J Emerg Med 2007;25(5):56471.
5. Cheng AC, West TE, Peacock SJ. Surviving sepsis in developing
countries. Crit Care Med 2008;36(8):2487, author reply 2487-8.
6. Cheng AC et al. Strategies to reduce mortality from bacterial
sepsis in adults in developing countries. PLoS Med 2008;5(8):e175.
7. Baelani I et al. Availability of critical care resources to treat
patients with severe sepsis or septic shock in Africa: a
134
self-reported, continent-wide survey of anaesthesia providers. Crit
Care 2011;15(1):R10.
8. Dellinger RP et al. Surviving sepsis campaign: international
guidelines for management of severe sepsis and septic shock:
2012. Crit Care Med 2013;41(2):580637.
9. Zambon M et al. Implementation of the surviving sepsis cam-
paign guidelines for severe sepsis and septic shock: we could go
faster. J Crit Care 2008;23(4):45560.
10. Jawad I, Luksic I, Rafnsson SB. Assessing available information
on the burden of sepsis: global estimates of incidence, prevalence
and mortality. J Global Health 2012;2(1):10404.
11. Dunser MW et al. Recommendations for sepsis management in
resource-limited settings. Intensive Care Med 2012;38(4):55774.
12. Black RE et al. Global, regional, and national causes of child
mortality in 2008: a systematic analysis. Lancet
2010;375(9730):196987.
13. Li J et al. Sepsis: the inammatory foundation of pathophysiol-
ogy and therapy.. Hosp Pract (1995) 2011;39(3):99112.
14. Dombrovskiy VY et al. Facing the challenge: decreasing case
fatality rates in severe sepsis despite increasing hospitalizations.
Crit Care Med 2005;33(11):255562.
15. Cohen J et al. Diagnosis of infection in sepsis: an evidence-based
review. Crit Care Med 2004;32(11 Suppl.):S46694.
16. Maki DG, Kluger DM, Crnich CJ. The risk of bloodstream
infection in adults with different intravascular devices: a system-
atic review of 200 published prospective studies. Mayo Clin Proc
2006;81(9):115971.
17. Marshall JC et al. Source control in the management of severe
sepsis and septic shock: an evidence-based review. Crit Care Med
2004;32(11 Suppl.):S51326.
18. Martin GS et al. The epidemiology of sepsis in the United States
from 1979 through 2000. N Engl J Med 2003;348(16):154654.
19. Lamy B et al. What is the relevance of obtaining multiple blood
samples for culture? A comprehensive model to optimize the
strategy for diagnosing bacteremia. Clin Infect Dis
2002;35(7):84250.
20. Shoemaker WC et al. Prospective trial of supranormal values of
survivors as therapeutic goals in high-risk surgical patients. Chest
1988;94(6):117686.
21. Rivers E et al. Early goal-directed therapy in the treatment of
severe sepsis and septic shock. N Engl J Med
2001;345(19):136877.
22. Jones AE et al. The effect of a quantitative resuscitation strategy
on mortality in patients with sepsis: a meta-analysis. Crit Care
Med 2008;36(10):27349.
23. ProCESS Investigators Yealy DM, Kellum JA, Huang DT Barnato
LA, Weissfeld LA, et al. A randomized trial of protocol-based care
for early septic shock. N Engl J Med 2014;370(18):168393.
24. Vincent JL, Gerlach H. Fluid resuscitation in severe sepsis and
septic shock: an evidence-based review. Crit Care Med 2004;32(11
):S4514.
25. Marik PE, Monnet X, Teboul JL. Hemodynamic parameters to
guide uid therapy. Ann Intensive Care 2011;1(1):1.
26. Marik PE, Baram M, Vahid B. Does central venous pressure
predict uid responsiveness? A systematic review of the literature
and the tale of seven mares. Chest 2008;134(1):1728.
27. Nguyen HB et al. Outcome effectiveness of the severe sepsis
resuscitation bundle with addition of lactate clearance as a bundle
item: a multi-national evaluation. Crit Care 2011;15(5):R229.
28. IMAI district clinician manual: hospital care for adolescents and
adults. Guidelines for the management of illnesses with limited
resources, 2011. Geneva: World Health Organization; 2011.
29. Guidet B et al. Assessment of hemodynamic efcacy and safety of
6% hydroxyethyl starch 130/0.4 vs. 0.9% NaCl uid replacement
in patients with severe sepsis: the CRYSTMAS study. Crit Care
2012;16(3):R94.
30. Perner A et al. Hydroxyethyl starch 130/0.42 versus Ringers
acetate in severe sepsis. N Engl J Med 2012;367(2):12434.
K. Tupchong et al.
31. Yunos NM et al. Association between a chloride-liberal vs
chloride-restrictive intravenous uid administration strategy and
kidney injury in critically ill adults. JAMA 2012;308(15):156672.
32. Shaw AD et al. Major complications, mortality, and resource
utilization after open abdominal surgery: 0.9% saline compared to
Plasma-Lyte. Ann Surg 2012;255(5):8219.
33. Maitland K et al. Mortality after uid bolus in African children
with severe infection. N Engl J Med 2011;364(26):248395.
34. Hollenberg SM et al. Practice parameters for hemodynamic
support of sepsis in adult patients: 2004 update. Crit Care Med
2004;32(9):192848.
35. LeDoux D et al. Effects of perfusion pressure on tissue perfusion
in septic shock. Crit Care Med 2000;28(8):272932.
36. De Backer D et al. Comparison of dopamine and norepinephrine
in the treatment of shock. N Engl J Med 2010;362(9):77989.
37. Tsuneyoshi I et al. Hemodynamic and metabolic effects of low-
dose vasopressin infusions in vasodilatory septic shock. Crit Care
Med 2001;29(3):48793.
38. Hebert PC et al. A multicenter, randomized, controlled clinical
trial of transfusion requirements in critical care. Transfusion
Requirements in Critical Care Investigators, Canadian Critical
Care Trials Group. N Engl J Med 1999;340(6):40917.
39. Levy MM et al. The surviving sepsis campaign: results of an
international guideline-based performance improvement program
targeting severe sepsis. Crit Care Med 2010;38(2):36774.
40. Jones AE et al. Lactate clearance vs central venous oxygen
saturation as goals of early sepsis therapy: a randomized clinical
trial. JAMA 2010;303(8):73946.
41. Kumar A et al. Duration of hypotension before initiation of
effective antimicrobial therapy is the critical determinant of
survival in human septic shock. Crit Care Med
2006;34(6):158996.
42. Gaieski DF et al. Impact of time to antibiotics on survival in
patients with severe sepsis or septic shock in whom early goal-
directed therapy was initiated in the emergency department. Crit
Care Med 2010;38(4):104553.
43. Leibovici L et al. Monotherapy versus beta-lactam-aminoglyco-
side combination treatment for gram-negative bacteremia: a
prospective, observational study. Antimicrob Agents Chemother
1997;41(5):112733.
44. Ulldemolins M et al. Appropriateness is critical. Crit Care Clin
2011;27(1):3551.
45. Bochud PY et al. Antimicrobial therapy for patients with severe
sepsis and septic shock: an evidence-based review. Crit Care Med
2004;32(11 Suppl.):S495512.
46. Pappas PG et al. Clinical practice guidelines for the management
of candidiasis: 2009 update by the Infectious Diseases Society of
America. Clin Infect Dis 2009;48(5):50335.
47. Berkley JA et al. HIV infection, malnutrition, and invasive
bacterial infection among children with severe malaria. Clin Infect
Dis 2009;49(3):33643.
48. Auma MA et al. Malaria is an uncommon cause of adult sepsis in
south-western Uganda. Malar J 2013;12:146.
49. WHO. World malaria report, 2011. Geneva: World Health Orga-
nization; 2012.
50. Hendriksen IC et al. Evaluation of a PfHRP2 and a pLDH-based
rapid diagnostic test for the diagnosis of severe malaria in 2
populations of African children. Clin Infect Dis 2011;52(9):11007.
51. Mtove G et al. Use of an HRP2-based rapid diagnostic test to
guide treatment of children admitted to hospital in a malaria-
endemic area of north-east Tanzania. Trop Med Int Health
2011;16(5):54550.
52. Ndyomugyenyi R, Magnussen P, Clarke S. Diagnosis and
treatment of malaria in peripheral health facilities in Uganda:
ndings from an area of low transmission in south-western
Uganda. Malar J 2007;6:39.
53. Ndyomugyenyi R, Magnussen P, Clarke S. Malaria treatment-
seeking behaviour and drug prescription practices in an area of
Sepsis, severe sepsis, and septic shock
low transmission in Uganda: implications for prevention and
control. Trans R Soc Trop Med Hyg 2007;101(3):20915.
54. McCombie SC. Treatment seeking for malaria: a review of recent
research. Soc Sci Med 1996;43(6):93345.
55. Marshall JC, Al Naqbi A. Principles of source control in the
management of sepsis. Crit Care Nurs Clin North Am
2011;23(1):99114.
56. Rothwell PM, Udwadia ZF, Lawler PG. Cortisol response to
corticotropin and survival in septic shock. Lancet
1991;337(8741):5823.
57. Cronin L et al. Corticosteroid treatment for sepsis: a critical
appraisal and meta-analysis of the literature. Crit Care Med
1995;23(8):14309.
58. Annane D et al. Effect of treatment with low doses of hydrocor-
tisone and udrocortisone on mortality in patients with septic
shock. JAMA 2002;288(7):86271.
59. Sprung CL et al. Hydrocortisone therapy for patients with septic
shock. N Engl J Med 2008;358(2):11124.
60. Finfer S et al. Intensive versus conventional glucose control in
critically ill patients. N Engl J Med 2009;360(13):128397.
61. Sevransky JE, Levy MM, Marini JJ. Mechanical ventilation in
sepsis-induced acute lung injury/acute respiratory distress syn-
drome: an evidence-based review. Crit Care Med 2004;32(11
):S54853.
62. Dmello D et al. Outcomes of etomidate in severe sepsis and septic
shock. Chest 2010;138(6):132732.
63. Tekwani KL et al. A comparison of the effects of etomidate and
midazolam on hospital length of stay in patients with suspected
sepsis: a prospective, randomized study. Ann Emerg Med
2010;56(5):4819.
64. Jabre P et al. Etomidate versus ketamine for rapid sequence
intubation in acutely ill patients: a multicentre randomised
controlled trial. Lancet 2009;374(9686):293300.
135
65. Tekwani KL et al. A prospective observational study of the effect
of etomidate on septic patient mortality and length of stay. Acad
Emerg Med 2009;16(1):114.
66. Barbi E, Rizzello E, Taddio A. Use of ketamine continuous
infusion for pediatric sedation in septic shock. Pediatr Emerg Care
2010;26(9):68990.
67. Morris C et al. Anaesthesia in haemodynamically compromised
emergency patients: does ketamine represent the best choice of
induction agent? Anaesthesia 2009;64(5):5329.
68. Ventilation with lower tidal volumes as compared with traditional
tidal volumes for acute lung injury and the acute respiratory
distress syndrome. The Acute Respiratory Distress Syndrome
Network, . N Engl J Med 2000;342(18):13018.
69. Drakulovic MB et al. Supine body position as a risk factor for
nosocomial pneumonia in mechanically ventilated patients: a
randomised trial. Lancet 1999;354(9193):18518.
70. Bataar O et al. Nationwide survey on resource availability for
implementing current sepsis guidelines in Mongolia. Bull World
Health Organ 2010;88(11):83946.
71. Santhanam I et al. GAP between knowledge and skills for the
implementation of the ACCM/PALS septic shock guidelines in
India: is the bridge too far? Indian J Crit Care Med 2009;13(2):548.
72. Jacob ST, West TE, Banura P. Fitting a square peg into a round
hole: are the current Surviving Sepsis Campaign guidelines feasible
for Africa? Crit Care 2011;15(1):117.
73. Lee BW. Improving sepsis care in resource limited settings. Crit
Care Med 2012;40(7):22346.
74. Jacob ST et al. The impact of early monitored management on
survival in hospitalized adult Ugandan patients with severe sepsis:
a prospective intervention study. Crit Care Med 2012;40(7):20508.
75. Becker JU et al. Surviving sepsis in low-income and middle-
income countries: new directions for care and research. Lancet
Infect Dis 2009;9(9):57782.